AI-Enabled Plaque Assessments Help Cardiologists Identify High-Risk CAD Patients

Groundbreaking research has shown that a non-invasive, artificial intelligence (AI)-based analysis of cardiac computed tomography (CT) can predict severe heart-related events in patients exhibiting symptoms of coronary artery disease (CAD). The late-breaking results from the CONFIRM2 global multicenter study highlighted the key atherosclerotic features most strongly linked to major adverse cardiovascular events (MACE) and revealed that these features are more reliable than traditional clinical risk scores.

In this study, which involved 3,551 symptomatic patients from 18 sites across 13 countries, researchers at Leiden University Medical Center (Leiden, Netherlands) used cutting-edge AI technology to analyze coronary computed tomography angiography (CCTA) data. The analysis revealed that two specific factors were the strongest indicators of MACE: 1) % Diameter Stenosis and 2) Non-Calcified Plaque Volume. Additionally, the CONFIRM2 study demonstrated that coronary artery calcified plaque volume did not independently predict negative outcomes for patients. These findings suggest that AI-driven evaluations could play a significant role in shaping treatment decisions, improving patient outcomes, and reducing the incidence of cardiovascular events in individuals with CAD.

"The integration of AI in assessing coronary artery disease represents a transformative leap in our ability to predict and manage coronary heart disease-related events,” said Dr. James K. Min, Founder and CEO of Cleerly. “AI-QCT analysis of cardiac CT provides diameter stenosis percentages and an accurate view of non-calcified plaque volume based on millions of images. This research not only highlights the potential of AI to improve diagnostic accuracy but also the importance of early intervention in reducing the risk of serious cardiovascular events."